Molecular interactions of endogenous D-myo-inositol phosphates with

May 11, 1994 - Intracellular D-myo-Inositol 1,4,5-Trisphosphate Recognition Site1 ... Eleven D-wpo-inositol phosphates were synthesized and tested for...
0 downloads 0 Views 1MB Size
Download PDF
Biochemistry 1994,33, 11586-1 1597

11586

Molecular Interactions of Endogenous D-myo-InOSitOl Phosphates with the Intracellular D-myo-Inositol 1,4,5-Trisphosphate Recognition Site$ Pei-Jung L U , ~ Da-Ming Gou,S Woan-Ru Shieh, and Ching-Shih Chen' Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, University of Rhode Island, Kingston, Rhode Island 02881 Received May 1 1 , 1994; Revised Manuscript Received July 28, 1994"

A systematic effort was made to elucidate the mode of recognition a t the inositol 1,4,5trisphosphate-specific receptor. Eleven D-myo-inositol phosphates were synthesized and tested for Ca2+mobilizing and receptor-binding activities, which included Ins( 1,3,4,5,6)P5, Ins( 1,2,5,6)P4, Ins( 1,3,4,5)P4, Ins( 1,3,4,6)P4, Ins( 1,4,5,6)P4, Ins(3,4,5,6)P4, Ins( 1,3,4)P3, Ins( 1,4,5)P3, Ins( 1,5,6)P3, Ins( 1,4)P2, and Ins(4,5)P2. Of these, Ins( 1,4,5)P3, Ins( 1,3,4,6)P4, Ins( 1,3,4,5)P4, Ins( 1,4,5,6)P4, and Ins(4,5)P2 were able to elicit Ca2+ release from rat brain microsomes. Binding experiments suggest that the ability of these polyphosphates to effect Ca2+mobilization arises from interactions with the Ins( 1,4,5)P3-specific receptor. Accordingly, a model accounting for the ligand recognition is proposed. The Ins( 1,4,5)P3-binding site is presumably composed of two domains. The anchoring domain binds the 4,5-bisphosphate 6-hydroxy motif. Disruption of this structural feature abolishes the agonist activity. The auxiliary domain exerts long-range interactions with the l-phosphate, thus enhancing the binding affinity. The stereochemical requirement for this electrostatic interaction is, however, less stringent. Evidence suggests that Ca2+-mobilizing inositol phosphates are able to effect productive binding by assuming conformations displaying or mimicking these essential structural features. ABSTRACT:

D-myo-Inositol 1,4,5-trisphosphate [Ins( 1,4,5)P,'] is released from the phospholipase C-mediated breakdown of phosphatidylinositol 4,5-bisphosphate [PI(4,5)P2] and plays an obligatory role in stimulating intracellular Ca2+mobilization (Berridge, 1993; Irvine, 1992a; Putney, 1992). It is well recognized that an elevation in cytosolic [Ca2+]signals diverse cellular responses, including cell growth and development, fertilization, secretion, smooth muscle contraction, sensory perception, and neuromodulation. This signal-transducing pathway is controlled by a multitude of mechanisms in a concerted manner (Marshall & Taylor, 1993). Evidence indicates that a heterogeneous but structurally related population of Ins( 1,4,5)P3receptors is differentially expressed in different tissues, as well as at different stages of development (Ferris & Snyder, 1992). Moreover, Ins( 1,4,5)P3,after being released into the cytosol, is rapidly metabolized by two discrete pathways mediated by a 5-phosphatase and by a 3-kinase to form Ins( 1,4)P2 and Ins( 1,3,4,5)P4, respectively (Fisher et al., 1992; Shears, 1992). While the formation of Ins(l,4)P2 represents a "switch-off" mechanism for Ins( 1,4,5)P3, the metabolism of Ins( 1,3,4,5)P4 and its metabolites by the sequential actions of specific phosphatases and kinases produces a plethora of inositol phosphates (Menniti et al., t

This paper is dedicated to Professor Charles J. Sih on the occasion

of his 60th birthday.

* Correspondingauthor: telephone, (401) 792-5048; Fax, (401) 7922181. s These coauthors contributed equally to this study. Abstract published in Advance ACS Abstracts, September 1 , 1994. Abbreviations: Ins( 1,4,5)P3, 1-D-myo-inositol 1,4,5-trisphosphate (other inositol phosphates are abbreviated similarly); PI(4,5)P2, phosphatidyl-D-myo-inositol4,S-bisphosphate; ee, enantiomeric excess; TLC, thin-layer chromatography; NMR, nuclear magnetic resonance; DMF, dimethylformamide; DTT, dithiothreitol; PMSF, p-toluenesulfonyl fluoride; EDTA, ethylenediaminetetraacetic acid; ECso, concentration causing half-maximal effect; ICso, concentration causing 50% displacement of specific binding. @

1993) (Scheme 1). These inositol phosphate congeners undergo dynamic turnover, and their cellular levels are sensitive to agonist stimulation. Evidently, the rich diversity of phosphoinositols generated from such a complex metabolic network implies the physiological relevance of these molecules. Some of these inositol phosphates have been reported to exhibit interesting biochemical activities. Ins( 1,3,4,5)P4has been shown to mobilize Ca2+ from the internal stores through interactions with the Ins( 1,4,5)P3-specificreceptor (Wilcox et al., 1993), and many studies have implicated Ins( 1,3,4,5)P4 in the regulation of Ca2+influx across the plasma membrane (Morris et al., 1987; Putney et al., 1989;Ely et al., 1990;Guseet al., 1992;Luckhoff et al., 1992; Irvine, 1992b; Smith, 1992). Ins(l,4)P2 has been reported to exert allosteric activation of muscle-type 6-phosphofructo-l-kinase (Mayr, 1989). It has been demonstrated that Ins(4,5)P2 and Ins(1,4,5)P3, but not Ins( 1,3,4,5)P4, selectively inhibited the Ca2+-ATPaseof rat heart sarcolemma (Kuo, 1988) and of human erythrocyte membrane (Davis et al., 1991). Ins( 1,3,4,6)P4-activated Ca2+ mobilization has been observed in microinjected Xenopus oocytes (Ivorra et al., 1991) and in permeabilized human neuroblastoma cells (Gawler et al., 1991). Moreover, it has been demonstrated that Ins(1,3,4,5,6)P~and IP6 were inhibitors of Ins(1,3,4,5)P4 3-phosphatase and Ins( 1,4,5)P3/Ins( 1,3,4,5)P4 5-phosphatase (Hoer & Oberdisse, 1991) and that Ins(1,3,4,6)P4 was a potent inhibitor of Ins( 1,4,5,6)P4 3-kinase (Craxton et al., 1994). The biological importance of Ins(1,4,5,6)P4 is indicated by the recent reports that intracellular Ins( 1,4,5,6)P4 levels were elevated upon receptor-coupled activation of phospholipase C (Barker et al., 1992) or subsequent to cell transformation with v-src oncogene (Mattingly et al., 1991). Despite recent advances in the synthesis of phosphoinositols [before 1990,see reviews by Potter (1990), Reitz (Ed.) (1991), Lin et al. (1990), Bruzik and Tsai (1992), Gou et al. (1992), Kozikowskietal. (l990,1993a),andPrestwichetal.(1991)], one major obstacle to investigating the biochemical functions

0006-2960/94/0433-11586$04.50/0 0 1994 American Chemical Society

Biochemistry, Vol. 33, No. 38, 1994

Inositol Phosphate Recognition at IP3 Receptor

11587

Scheme 1: Metabolism of Inositol Phosphates" OH OC(0)R

? O&OC(0)R

2'O.po

~-

0-p-

OH

II

0

PIP,

I

OH-

OH

3H

OH

OH

Intracellular Inositol Pool The mechanismof Ins(4,5)P2formation remains unclear. It may result from phospholipaseD-mediatedhydrolysis of PIP2 or from dephosphorylation of Ins(1,4,5)P3 or Ins(1,3,4,5)Pd (Jenkinson et al., 1992).

of these polyphosphates has been the difficulty in gaining access to these compounds in high optical and chemical purity. Commercial preparations have been reported to be contaminated with other active inositol phosphates (Parker & Ivorra, 1991), and their cost often becomes a prohibitive factor for in-depth studies. Moreover, it should be noted that the use of racemic inositol phosphates may complicate the interpretation of biochemical data. For example, DL-myo-inositol 1,4,5,6-tetrakisphosphateis a mixture of Ins( 1,4,5,6)P4 and Ins(3,4,5,6)P4, both of which are natural inositol phosphates. Our current study shows that while Ins( 1,4,5,6)P4 is active in eliciting Ca2+ release from internal stores, Ins(3,4,5,6)P4 is not. Thus, as part of our research on the physiological significance of this intricate metabolism, we embarked on a systematic synthesis of naturally occurring D-myo-inositol phosphates. Since each compound was independently synthesized and fully characterized, the possibility of cross-contamination by other phosphoinositols was excluded. With these compounds in hand, we initiated a series of examinations of their roles in regulating cytosolic [Ca2+]. In this account, we report the syntheses and Ca2+-mobilizingactivity of 11 D-myo-inositol phosphates. Accordingly, a working model accounting for the mode of interactions between endogenous inositol phosphates and the Ins( 1,4,5)P3 recognition site is proposed.

EXPERIMENTAL PROCEDURES General Methods and Reagents. 'H, I3C, and 31PN M R spectra were recorded with a Brucker AM-300 spectrometer. Optical rotations were determined with a Rudolph Autopol I11 polarimeter. HPLC was performed using a Model 501 pump (Waters Associates) equipped with a Rheodyne injector and a Model 481 UV/vis detector (Waters Associates). Elemental analysis was conducted by M-H-W laboratories (Phoenix, AZ). Fluorescence spectrophotometric assay of

Ca2+ release was carried out with a Hitachi F-2000 spectrometer. Racemic 1,2:5,6-di-Ocyclohexylidene-myo-inositol (2) and racemic 1,2:4,5-di-O-cyclohexylidene-myo-inositol (32) were prepared according to the procedures reported by Garegg et al. (1984). Ins( 1,4,5)P3, Ins( 1,3,4)P3, and Ins(1,3,4,5)P4 were synthesized from optically active 2 according to the procedures previouslyreported (Gou et al., 1992). [3H]Ins( 1,4,5)P3 (2 1 Ci/mmol) was purchased from DuPont New England Nuclear. Enzymatic Resolution of (*)-I ,2:5,6-Di-O-cyclohexylidene-myo-inositol(2). Optically active 2 was prepared by subjecting (&)-1 [for preparation, see Gou et al. (1992)l to enantioselective hydrolysis by porcine pancreatic lipase (PPL, Sigma type 11) in a biphasic system. A solution of the substrate (14 g) in hexane-ether (lO:l, 110 mL) was introduced into 0.1 M potassium phosphate buffer (pH 7.0,60 mL) containing crude PPL powder (40 g). The mixture was stirred vigorously at 37 OC, and the reaction was monitored by TLC. After 42 h, the reaction was terminated by separating the two layers. The organic layer was dried and concentrated. Column chromatography (hexane-ethyl acetate, 15:l 4:l) of the residue gave (-)-2 (5.4 g, 94% ee), [(Y]D-17.8'), and (-)-I (6.9 g, 87% ee, [(Y]D-13.2O). The remaining substrate was subjected to saponification (1 M NaOH-CH3OH, 23 OC, 3 h) to afford (+)-2. The optical purity of (+)- and (-)-2 was no less than 98% ee after recrystallization. The enantiomeric purity of 2 was determined by HPLC analysis of the corresponding(S)-2-methoxy-2-(trifluoromethyl)phenylacetyl ester, as previously reported (Gou et al., 1992). With (+)and (-)-2, 11 different D-myo-inositol phosphates were synthesized, including Ins( 1,3,4,5,6)P5, Ins( 1,2,5,6)P4, Ins(1,3,4,5)P~,Ins(1,3,4,6)P~,Ins(1,4,5,6)P~,Ins(3,4,5,6)P~,Ins(1,3,4)P3, Ins( 1,4,5)P3, Ins( 1,5,6)P3, Ins( 1,4)Pz, and Ins(4,5)P2. Syntheses of Ins(1,4,5)P3 from (+)-2 (Liu & Chen, 1989) and of Ins( 1,3,4)P3and Ins( 1,3,4,5)P4from (-)-2 (Gou

-

11588 Biochemistry, Vol. 33, No. 38, 1994 & Chen, 1992;Gou et al., 1992) have been described elsewhere and will be not elaborated here. (-)- 1,6- Di- 0-benzyl-2,3:4,5-di- 0-cyclohexylidene-myoinositol (3). A solution of (+)-2 (0.4 g, 1.1 mmol) in DMF (10 mL) was treated with N a H (100 mg, 80%, 3.3 mmol) and benzyl bromide (0.32 mL, 2.4 mmol) for 12 h at 35 'C. The excess NaH was destroyed with CH3OH, and the mixture wasdilutedwithethylacetate (50mL). After aqueousworkup, column chromatography (hexane-ether, 25:l) of the residue afforded (-)-3 (syrup, 0.6 g, 98%): [(u]D-21 '(c 1.4, CHC13); 'H NMR (CDC13) 6 1.27-1.78 (m, 20 H), 3.51 (dd, 1 H, J = 7.86,9.65 Hz), 3.76 (t, 1 H, J = 6.60 Hz), 3.87 (dd, 1 H, J = 2.91, 7.80 Hz), 4.13 (dd, 1 H, J = 3.76, 10.6 Hz), 4.31 (t, 1 H,7.28Hz),4.38 (dd, 1 H , J = 3.82,6.86Hz),4.50-4.73 (m, 4 H), 7.26-7.38 (m, 10 H). (+)- 1,6-Di-O-benzyI-2,3- 0-cyclohexylidene-myo-inosito1 (4). A solution of (-)-3 (0.5 g, 0.96 mmol) in CH2C12CH30H (3:1, 20 mL) was stirred with acetyl chloride (80 pL) at 23 OC. TLC analysis indicated that, under these conditions, the trans-cyclohexylidene function was totally eliminated within 10-15 min, while the cleavage of the cisketal ring was not appreciably noted until after 20 min. Thus, the solution was stirred for 10 min, triethylamine (0.3 mL) was then added, and the solution was concentrated. Column chromatography (hexane-ether, 8:1) of the residue yielded (+)-4 (Syrup, 330 mg, 78%): [ a ]+17' ~ (C 0.6, CHCl3); ' H NMR (CDC13) 6 1.26-1.78 (m, 10 H), 2.93 (br s, 2 H), 3.29 (dd, 1 H, J = 8.52, 10.1 Hz), 3.64-3.77 (m, 3 H), 3.88 (dd, 1 H, J = 5.27, 7.48 Hz), 4.27 (dd, 1 H, J = 3.88, 5.18 Hz), 4.74 (q,2 H, J = 12.2, 13.7 Hz), 4.84 (dd, 2 H, J = 11.2,77.2 Hz), 7.26-7.40 (m, 10 H). (+)-I ,6-Di- 0-benzyl-2,3- 0-cyclohexylidene-myo-inosito1 4,5-Bis(dibenzylphosphate) (5). A mixture of tetrazole (0.6 g, 8 mmol), dibenzyl N,N-diisopropylphosphoramidite (1.83 g, 4 mmol), and CH2Cl2 (20 mL) was stirred under Ar at 23 "C for 1 h, and (+)-4 was added in one portion. The solution was kept under the same conditions for another 12 h, cooled to -40 OC, and then treated with triethylamine (2 mL) and m-chloroperoxybenzoic acid (50% purity, 2 g, 4.1 mmol). The mixture was stirred at -40 'C for 30 min, then allowed to attain room temperature, diluted with CH2C12 (20 mL), and subjected to aqueous workup. The product was purified by column chromatography (hexane-ethyl acetate, 20:l 2:l) to give (+)-5 (syrup, 0.65 g, 85%): [(Y]D+2.3' (c 1.3, CHC13); 'H NMR (CDC13) 6 1.27-1.80 (m, 10 H), 3.79 (dd, 1 H, J = 3.74, 7.39 Hz), 4.03 (dd, 1 H, J = 6.13,

-

7.36Hz),4.2O(t,lH,J=6.83Hz),4.35(dd,lH,J=3.81, 6.43 Hz), 4.49-4.57 (m, 1 H), 7.09-7.36 (m, 30 H). 1-D-myo-Inositol IS-Bisphosphate [Ins(4,5)P~].A solution of (+)-5 (0.65 g, 0.5 mmol) in aqueous 85% EtOH (30 mL) was shaken under H2 (50 psi) in the presence of 10% Pd/C (0.4 g) for 24 h, filtered, and concentrated. The residue was dissolved in minimal water, and 4 equiv of 1 M KOH was added. The solution was lyophilized to afford Ins(4,5)P2 as the tetrapotassium salt (0.28 g, 99%): [a]D -10' (c 1, H20); 'H NMR (D20) 6 3.42-3.46 (m, 1 H), 3.54 (dd, 1 H, J = 3.45, 9.60 Hz), 3.67-3.74 (m, 2 H), 3.87 (t, 1 H, J = 2.83 Hz), 3.99 (m, 1 H); 3 1 P N M R(D20, external H3P04) 6 4.62, 4.77. I-) - I ,6Di- O-acetyl-2,3:4,5-di- 0-cyclohexylidene-myoinositol ( 6 ) . A solution of (+)-2 (0.5 g, 1.4 mmol) in CHzCl2 (10 mL) was treated with acetic anhydride (1.2 mL), triethylamine (2 mL), and 4-(dimethy1amino)pyridine (100 mg) at 23 'C for 30 min. After the aqueous workup, column chromatography (hexane-ether, 20:l 5:l) of the residue gave (-)4 (syrup, 0.6 g, 96%): [(]LID -5.3' (c 1, CHC1,); lH

-

Lu et al. NMR(CDC13)J 1.26-1.88(m,20H),2.10(s,3H),2.12(s, 3 H), 3.59-3.65 (m, 1 H), 4.02-4.09 (m, 1 H), 4.38 (t, 1 H, J = 7.58 Hz), 4.44-4.51 (m, 1 H), 5.15-5.21 (m, 2 H). I-) -1,6- Di- O-acetyl-2,3- 0-cyclohexylidene-myo-inosito1 (7). The foregoing fully protected myo-inositol(-)-6 (0.6 g, 1.4 mmol) in CH2C12-CH30H (2:1, 15 mL) was treated with acetyl chloride (50 pL) to remove the 4,5-transcyclohexylidene ring, as described for 4, to furnish (-)-7 (0.43 g, 80%): [a]D -25' (C 0.7, CHCl3); 'H NMR (CDC13) 6 1.35-1.68 (m, 10 H), 2.10 (s, 6 H), 3.05 (br s, 2 H), 3.46 (t, 1 H, J = 8.04 Hz), 3.82 (dd, 1 H, J = 7.39, 9.91 Hz), 4.08 (dd, 1 H, J = 5.61, 7.30 Hz), 4.44 (dd, 1 H, J = 3.54, 5.57 Hz), 5.10-5.52 (m, 2 H). (+)-2,3-0-Cyclohexylidene-myo-inositol(8).A mixture of (-)-7 (390 mg, 1 mmol), KOH (5.6 mg, 0.1 mmol), and CH30H (10 mL) was stirred at 23 OC for 30 min. After aqueous workup, the residue was subjected to column chromatography (hexane-ethyl acetate, 15:1 2: 1) to afford (+)-8(solid, 290 mg, 97%): mp 188-189 'C; [(Y]D+35' (c 1.4, CH30H); ' H NMR (CD30D) 6 1.36-1.66 (m, 10 H), 3.07 (dd, 1 H, J = 9, 10 Hz), 3.44-3.57 (m, 2 H), 3.59-3.64 (m, 1 H), 3.88 (dd, 1 H, J = 4.99, 7.51 Hz), 4.31 (t, 1 H, J = 4.94 Hz). Anal. Calcd for C12H2006: C, 55.37; H, 7.74. Found: C, 55.77; H, 7.59. (-) - 2,3- 0-Cyclohexylidene- myo- inositol I ,4,5,6- Tetra kis(dibenzy1 phosphate) ( 9 ) . The tetrol (+)-8 (280 mg, 1 mmol) was phosphorylated with dibenzyl N,N-diisopropylphosphoramidite, tetrazole, and m-chloroperoxybenzoicacid, asdescribedfor 5, to yield (-)-9 (syrup, 1 g, 77%): [a]~-8.5' (c2, CHC13); IH NMR (CDC13) 6 1.25-1.71 (m, 10H),4.26 (t, 1 H, J = 5.88 Hz), 4.68 (dd, 1 H, J = 3.62, 6.68 Hz), 4.70-4.80 (m, 2 H), 4.91-5.12 (m, 16 H), 7.18-7.29 (m, 40 H). 1-D-myo-Inositol I ,4,5,6- Tetrakisphosphate [Zns(l,4,5,6)P41. A mixture of (-)-9 (780 mg, 0.5 mmol) and 10% Pd/C (250 mg) in aqueous 85% EtOH (30 mL) was shaken under H2 (50 psi) for 12 h, filtered, and concentrated. The residue was dissolved in the minimum amount of water, 1 M KOH (8 equiv) was added, and the mixture was lyophilized to afford Ins( 1,4,5,6)P., as the octapotassium salt (0.47 g, 99%): [(Y]D -6.9' (C 0.65, H20); 'H N M R (D2O) 6 3.95 (t, 1 H, J = 3.6 Hz), 4.04-4.10 (m, 2 H), 4.37 (d, 2 H, J = 10.8 Hz), 4.58 (d, 1 H, J = 10.2 Hz); 31PN M R (D20, external H3P04) 6 4.25, 4.56, 5.18, 5.33. (-) - 4- 0-Allyl- 1 , 2 5 6 - di- 0-cyclohexylidene- myo- inosit o l ( l 0 ) . Amixtureof(-)-2(1.2g, 3.5mmol),BuzSnO(0.98 g, 3.8 mmol), and toluene (30 mL) was boiled under reflux, with azeotrophic removal of water for 1 h, and then concentrated to dryness under reduced pressure. To the residue were added DMF (10 mL), CsF (1.3 g, 8.2 mmol), and allyl bromide (0.72 mL, 8.2 mmol). The mixture was stirred under Ar at 23 'C for 16 h and then diluted with ethyl acetate (80 mL). After aqueous workup, column chromatography (hexane-ether, 20:l 1O:l) gave (-)-lo (syrup, 0.85 g, 63%): [a]D -4.5' (C 0.8, CHCl3); 'H N M R (CDC13) 6 1.40-1.72 (m, 20 H), 3.49 (dd, 1 H, J = 7.91, 10.6 Hz), 3.84 (dd, 1 H, J = 1.92,7.87 Hz), 3.98-4.00 (m, 1 H), 4.10-4.28 (m, 3 H), 4.34 (t, 1 H, J = 6.6 Hz), 4.43 (dd, 1 H, J = 3.64,7.38 Hz), 5.21 (dq, 1 H, J = 1.2,2.9, 10.4 Hz), 5.33 (dq, 1 H, J = 1.57, 3.27, 17.2 Hz), 5.87-6.01 (m, 1 H). (+)-4- 0-Allyl-3- 0-benzyl- 1,2:5,6-di- 0-cyclohexylidenemyo-inositol(11). Benzylation of (-)-lo (0.83 g, 2.1 mmol) with benzyl bromide, as described for 3, furnished (+)-ll (syrup, 1 g, 99%): [ a ] +12.5' ~ (c 0.55, CHC13); 'H NMR (CDC13) 6 1.39-1.75 (m, 20 H), 3.43 (dd, 1 H, J = 7.7, 10.6 Hz), 3.71 (t, 1 H , J = 3.28 Hz), 3.79 (dd, 1 H , J = 2.88,7.76

-

-

Inositol Phosphate Recognition at IP3 Receptor Hz), 3.94-4.01 (m, 1 H), 4.05-4.14 (m, 2 H), 4.25-4.37 (m, 2 H), 4.72 (q, 1 H, J = 12.4, 19.3 Hz), 5.16 (dq, 1 H, J =

Biochemistry, Vol. 33, No. 38, I994

11589

p-toluenesulfonic acid monohydrate (100 mg, 0.52 mmol) in aqueous 90% EtOH (20 mL) was stirred under reflux for 45 1.45,3.11,10.3Hz),5.26(dq,1H,J=1.55,3.85,17.3Hz), min and then filtered and concentrated. Column chroma5.80-5.93 (m, 1 H), 7.27-7.40 (m, 5 H). tography (hexane-ethyl acetate, 21:l 5:l) of the residue afforded (+)-17 (syrup, 240 mg, 69%): [a]D +5' (c 0.4, (-)-4- 0-Allyl-3- 0-benzyl- I ,2- 0-cyclohexylidene-myoCHC13); 'H N M R (CDC13) 6 2.30 (d, 1 H, J = 6.0 Hz), 2.55 inositol ( 12). Regioselective hydrolysis of the trans-cyclo(d, 1 H, J = 1.5 Hz), 3.33 (dd, 1 H, J = 2.3, 9.8 Hz), 3.39 hexylidene of (+)-11 (0.7 g, 1.5 mmol), as described for 4, (t, 1 H, J = 9.1 Hz), 3.48-3.54 (m, 1 H), 3.78 (t, 1 H, J = gave (-)-12 (syrup, 0.45 g, 77%): [a]D -9' (c 0.6, CHC13); 9.3 Hz), 4.06 (t, 1 H, J = 2.4 Hz), 4.15 (t, 1 H, J = 9.2 Hz), 'H NMR (CDC13) 6 1.32-1.78 (m, 10 H), 2.73 (br s, 2 H), 4.57-4.78 (m, 4 H), 4.86-4.93 (m, 4 H), 7.26-7.45 (m, 20 3.22-3.32 (m, 1 H), 3.62-3.64 (m, 2 H), 3.74 (dd, 1 H, J = 7.58, 10.1 Hz), 3.91 (dd, 1 H, J = 5.21, 7.5 Hz), 4.18-4.29 H) * (m, 2 H), 4.42-4.49 (m, 1 H), 4.75 (q, 2 H, J = 11.5, 13.7 (+)- 2,3,5,6- Tetra- 0-benzyl-myo-inositol I ,4-Bis(dibenHz), 5.18-5.34 (m, 2 H), 5.91-6.03 (m, 1 H), 7.30-7.41 (m, zylphosphate) (18). The diol (+)-17 (170 mg, 0.32 mmol) 5 H); I3CNMR (CDC13) 6 23.64,23.94,25.05,35.17,37.76, was phosphorylated, as described for 5, to give (+)-18 (syrup, 65.86, 72.74, 73.18, 73.65, 73.83, 75.2, 77.23, 77.85, 78.23, 300mg, 89%): [CY]D +5.8' (C 1, CHC13); 'H NMR (CDCl3) 80.1, 110.73, 117.21, 127.96, 128.02, 128.48, 134.9, 138.05. 6 3.39-3.43 (m, 1 H), 3.50 (t, 1 H, J = 9.2 Hz), 4.07 (t, 1 H, (-)-4-O-Allyl-3,5,6-tri-O-benzyl-1,2-O-cyclohexylidene- J = 9.4 Hz), 4.24-4.27 (m, 1 H), 4.32 (m, 1 H), 4.52 (s, 2 myo-inositol (13). Benzylation of (-)-12 (0.4 g, 1 mmol) H), 4.61-4.99 (m, 15 H), 7.03-7.31 (m, 40 H). with benzyl bromide, as described for 3, yielded (-)-13 (syrup, I -D-myo-Inositol I,4-Bisphosphate [Zns(l,4)P2]. The 0.52 g, 90%): [a]D -18' (C 1.2, CHCl3); 'H NMR (CDC13) foregoing compound (+)-18 (250 mg, 0.23 mmol) was 6 1.37-1.78 (m, 10 H), 3.34 (dd, 1 H, J = 8.42,9.6 Hz), 3.62 debenzylated by hydrogenolysis, followed by KOH titration (dd, 1 H, J = 3.76, 8.11 Hz), 3.73-3.82 (m, 2 H), 4.07 (dd, and lyophilization, as described for Ins(4,5)P2, to generate 1 H, J = 5.6, 6.97 Hz), 4.23-4.37 (m,3 H), 4.71-4.90 (m, Ins( 1,4)P2 as the tetrapotassium salt (1 15 mg, 99%): [&ID 6 H), 5.14-5.19 (m, 1 H), 4.24-5.32 (m, 1 H), 5.91-6.05 (m, +3.6' (C0.5,H2O); 'H NMR (D2O) 6 3.32 (t, 1 H,J = 9.3 1 H), 7.23-7.42 (m, 15 H); 13CN M R (CDC13) 6 23.6,23.88, Hz),3.52(dd,lH,J=3,6.45Hz),3.67(t,lH,J=3.9Hz), 25.04,35.02, 37.33,73.07,73.63,73.89,75.16,77.09,78.75, 3.75-3.82 (m, 1 H), 4.0 (dd, 1 H, J = 7.61, 11.3 Hz), 4.08 80.57,82.13,82.80, 110.36, 116.57, 127.38 (2C),127.47 (2 (t, 1 H, J = 2.67 Hz); 31PNMR (D20, external H3P04) 6 C), 127.64 (2 C), 127.83 (2 C), 127.98 (2 C), 128.17 (2 C), 3.12, 4.03. 128.2 (2 C), 128.28 (2 C), 135.16, 138.39, 138.67, 138.72. (+) - 3,4- Di- 0-benzyl- 1,2:5,6-di- 0-cyclohexylidene-myo(-)-4- 0Allyl-3,5,6-tri- 0-benzyl-myo-inositol(l4). The inositol (19). Conventional benzylation of (-)-2 (1 g, 2.9 foregoing fully protected inositol (-)-13 (0.5 g, 0.87 mmol) mmol), as described for 3, afforded (+)-19 (1.4 g, 90%): [a]D was treated with acetyl chloride (50 ML)in CH2C12-CH3OH +21° (c 1, CHC13); 'H NMR (CDCl3) 6 1.27-1.78 (m, 20 ( l : l , 25 mL) at 23 'C for 30 min, and the solution was H), 3.51 (dd, 1 H, J = 7.86,7.95 Hz), 3.76 (t, 1 H, J = 6.60 concentrated. Column chromatography (hexane-ether, 20: 1 Hz),3.87(dd,lH,J=2.9,7.8Hz),4.13(dd,lH,J=7.36, 1O:l) of the residue gave (-)-14 (syrup, 0.4 g, 93%): [a]D 10.6 Hz), 4.31 (t, 1 H, J = 7.28 Hz), 4.38 (dd, 1 H , J = 3.81, -8.3' (C 0.6, CHC13); 'H N M R (CDC13) 6 2.43 (d, 1 H, J 6.86 Hz), 4.50-4.73 (m, 4 H), 7.26-7.38 (m, 10 H). = 4.16 Hz), 2.48 (s, 1 H), 3.25-3.41 (m, 4 H), 3.37-3.47 (m, (-) - 3,4- Di- 0-benzyl- 1,2-0-cyclohexylidene- myo-inosi3 H), 3.80 (dd, 2 H, J = 9.47, 19.4 Hz), 4.17 (t, 1 H, J = to1 (20). Hydrolysis of the trans-cyclohexylidene ring of (+)2.84 Hz), 4.29-4.43 (m, 2 H), 4.66-4.82 (m, 4 H), 4.92 (t, 2 H , J = 10.6Hz),5.14-5.19(m,lH),5.25-5.32(m,lH),19 (1.2 g, 2.3 1 mmol), as described for 4, yielded (-)-20 (syrup, 0.7 g, 70%): [a]D -16.5' (C 1, CHC13); 'H NMR (CDCl3) 5.92-6.05 (m, 1 H), 7.26-7.37 (m, 15 H); I3CNMR (CDC13) 669.4,71.8,72.8,74.6,75.6,75.7,77.3,79.9,81.3,81.4,83.2,6 1.27-1.78 (m,10 H), 2.93 (br s, 2 H), 3.29 (dd, 1 H, J = 8.51, 10.1 Hz), 3.64-3.77 (m, 3 H), 3.88 (dd, 1 H, J = 5.27, 116.6, 127.6, 127.8 (3 C), 127.9 (4 C), 128.4 (2 C), 128.5 (3 7.49 Hz), 4.27 (dd, 1 H, J = 3.88, 5.18 Hz), 4.74 (q, 2 H, C), 128.6 (2 C), 135.3, 137.9, 138.6. J =12.2,13.7H~),4.84(dd,2H,J=11.2,77.2H~),7.26(+)- 1,4- Di- 0-allyl- 3,5,6-tri- 0-benzyl-myo-inositol(l5). 7.40 (m, 10 H). Regioselective allylation of (-)-14 (0.4 g, 0.8 mmol), as described for 10, gave (+)-15 (syrup, 0.4 g, 94%): [a]D +7.8' (-) -5,6- Di- 0-allyl- 3,4-di- 0-benzyl- I ,2- 0-cyclohex(C0.5, CHCl3); 'H NMR (CDC13) 6 2.43 (s, 1 H), 3.25-3.41 ylidene-myo-inositol (21). Conventional allylation of (-)(m, 4 H), 3.80-3.93 (m, 2 H), 4.13-4.21 (m, 2 H), 4.29-4.42 20 (0.65 g, 1.47 mmol) with allyl bromide furnished (-)-21 (m, 2 H), 4.73 (q,2 H, J = 10.6, 15 Hz), 4.79-4.87 (m, 4 H), (syrup, 0.73 g, 95%): [a]D -26.2' (C 1.8, CHC13); 'H NMR 5.14-5.20(m,2H),5.24-5.31(m,2H),5.86-6.05(m,2H), (CDC13) 6 1.25-1.80 (m, 10 H), 3.20 (dd, 1 H, J = 8.71,9.22 7.26-7.41 (m, 15 H). Hz),3.58-3.64(m,2H),3.84(t,l H,J=8.35Hz),3.95(dd, 1 H, J = 5.46, 6.33 Hz), 4.21-4.37 (m, 5 H), 4.69-4.84 (m, (+)-I ,4- Di- 0-allyl- 2,3,5,6-tetra- 0-benzyl-myo-inosi4 H), 5.12-5.14 (m, 1 H), 5.16-5.18 (m, 1 H), 5.24-5.26 (m, to1 (16). Benzylation of (+)-15 (0.38 g, 0.67 mmol), as 1 H), 5.29-5.32 (m, 1 H), 5.89-6.0 (m, 2 H), 7.25-7.39 (m, described for 3, yielded (+)-16 (syrup, 0.42 g, 95%): [a]D 10 H). +5.7' (c 0.9, CHC13); 'H NMR (CDC13) 6 3.21-3.32 (m, 2 H),3.40(t, lH,J=9.26Hz),3.89-4.13(m,4H),4.26-4.43 (-)-5,6- Di- O-allyl-3,4-di- 0-benzyl-myo-inositol(22). (m, 3 H), 4.64 (9, 2 H, J = 11.8, 25.4 Hz), 4.76-4.90 (m, 6 Hydrolysis of the cis-cyclohexylidene of (-)-21 (0.7 g, 1.3 H), 5.1 1-5.33 (m, 4 H), 5.83-6.04 (m, 2 H), 7.22-7.42 (m, mmol), as described for 14, gave (-)-22 (syrup, 0.56 g, 95%): 20H);'3CNMR(CDC13)671.38,72.90,74.07,74.58,75.83, [a]D-15.S0 (c 0.9, CHC13); 'H NMR (CDC13) 6 2.32 (br s, 75.94, 77.23, 80.75, 80.87, 81.5, 81.65, 83.73, 116.51 (3 C), 2 H), 3.26 (t, 1 H, J = 9.40 Hz), 3.40 (dd, 1 H, J = 2.73, 116.63 (3 C), 127.31 (3 C), 127.52 (3 C), 127.82 (3 C), 127.95 9.4Hz), 3.65 (t, 1 H , J = 9.5 Hz), 3.86 (t, 1 H , J = 9.48Hz), (3 C), 128.12 (3 C), 128.32 (3 C), 135.01, 135.53, 138.62, 4.18-4.46 (m, 5 H), 4.70 (4,2 H, J = 11.6, 12.9 Hz), 4.83 139.04. (q, 2 H, J = 10.62, 12.5 Hz), 5.14-5.33 (m, 4 H), 5.89-6.03 (+)-2,3,5,6- Tetra-0-benzyl-myo-inositol(17). A mix(m, 2 H), 7.26-7.37 (m, 10 H); 13C NMR (CDC13) 6 69.4, ture of (+)-16 (0.4 g, 0.65 mmol), 10% Pd/C (100 mg), and 71.76, 72.84, 74.25, 74.3, 75.88, 77.19, 80.07, 80.95, 81.64,

-

-

11590 Biochemistry, Vol. 33, No. 38, 1994

Lu et al.

82.96, 116.43, 116.94, 127.55, 127.86, 127.88, 127.99 (2 C), (m, 1 H); 31PNMR(D20, externalH3PO.d 64.13,4.25,4.37, 128.31 (2 C), 128.5 (2 C), 135.22, 135.3, 138.01, 138.92. 4.50. (+)-3,4-Di- 0-acetyl- I ,2:5,6-di- O-cyclohexylidene-myo(-I- 1.5,6-Tri- O-allyl-3,4-di-0-benzyl-myo-inositol(23). inositol (29). Acetylation of (-)-2 (0.6 g, 1.6 mmol), as Regioselective allylation of (-)-22 (0.55 g, 1.2 mmol), as describedfor 6,gave (+)-29 (syrup, 0.74 g, 99%): [a]D +5.4' described for 10,yielded (-)-23 (syrup, 0.56 g, 93%): [a]D ( c 1.1, CHC13); 'H N M R (CDCl3) 6 1.26-1.87 (m, 20 H), -14.5' ( C 1.1, CHCl3); 'H N M R (CDC13) 6 2.43 (s, 1 H), 2.10 (s, 3 H), 2.12 (s, 3 H), 3.59-3.65 (m, 1 H), 4.02-4.09 3.15-3.26 (m, 2 H), 3.32-3.36 (m, 1 H), 3.73 (t, 1 H, J = 9.6Hz),3.87(t,l H,J=9.6Hz),4.17-4.18(m,3H),4.27- (m, 1 H), 4.38 (t, 1 H, J = 7.58 Hz), 4.44-4.51 (m, 1 H), 5.15-5.20(m, 2 H); 13CNMR(CDC13) 6 20.88,20.98,23.56, 4.32 (m, 3 H), 4.71 (d, 2 H, J = 3.0 Hz), 4.83 (s, 2 H), 23.69, 23.91, 25.04, 25.12, 25.36, 31.61, 34.32, 36.37, 36.7, 5.1 1-5.13 (m, 1 H), 5.15-5.17 (3, 1 H), 5.18-5.19 (m, 1 H), 36.79,72.38,73.62,75.87,76.33,77.27,111.89,113.66,169.1, 5.23-5.29 (m, 2 H), 5.29-5.31 (m, 2 H), 5.84-6.04 (m, 3 H), 169.39. 7.25-7.37 (m, 10 H). (-)-1,2-O-Cyclohexylidene-myo-inositol(30). A solution (+)- I S $ - Tri- O-allyl-2.3,4-tri- 0-benzyl-myo-inositol of (+)-29(0.7 g, 1.6 mmol) in CH2C12-CH30H (1:1,14 mL) (24). Conventional benzylation of (-)-23(0.55 g, 1.1 mmol), was stirred with acetyl chloride (60 pL, 0.72 mmol) at 23 'C as described for 3, gave (+)-24 (syrup, 0.6 g, 93%): [a]D for 5 min. Triethylamine (100 pL, 0.72 mmol) was added, +4.1' ( c 2.8, CHCl3); 'H N M R (CDCl3) 6 3.15-3.36 (m, 3 the solution was concentrated, and the residue was treated H), 3.96-3.97 (m, 1 H), 4.05-4.08 (m, 1 H), 4.14-4.18 (m, with methanolic 1 M KOH for 1 h at 23 'C. After aqueous 2 H), 4.28-4.33 (m, 4 H), 4.56-4.71 (m, 3 H), 4.81-4.85 (m, workup, column chromatography (hexaneethyl acetate, 2: 1) 3 H), 5.12-5.19 (m, 3 H), 5.23-5.26 (m, 2H), 5.30-5.31 (m, afforded (-)-30 (solid, 0.35 g, 78%): [a]D -35' (c 0.9, CH32 H), 5.92-5.98 (m, 3 H), 7.25-7.42 (m, 15 H). OH); 'H NMR (CD30D) 6 1.40-1.71 (m, 10 H), 3.11 (t, 1 (+) - 2,3,4- Tri- 0-benzyl-myo- inositol (25). Deallylation H, J = 9.31 Hz), 3.47-3.68 (m, 2 H), 3.59-3.64 (m, 1 H), of the foregoing fully protected inositol (+)-24 (0.58 g, 1 3.91 (dd, 1 H, J = 5.03, 7.46 Hz), 4.35 (t, 1 H, J = 4.46); mmol) with 10%Pd/C andp-toluenesulfonicacid,as described I3CN M R (CDC13) 6 24.61,24.94,26.11,36.21,39.13,7 1.80, for 17,afforded (+)-25(syrup, 0.4 g, 86%): [a]D +59.6' (c 73.86, 75.14, 77.02, 77.35, 80.25, 111.34. Anal. Calcd for 0.55, CHC13); 'H NMR (CDC13) 6 3.33-3.39 (m, 3 H), 3.44ClzH2oOa: C, 55.37; H, 7.74. Found: C, 55.77; H, 7.59. 3.47 (m, 2 H), 3.73 (t, 1 H, J = 9.6 Hz), 3.85 (t, 1 H, J = (+)-I , 2 - 0 - Cyclohexylidene-myo-inositol3,4,5,6Tetra9.3 Hz), 4.03 (br s, 1 H), 4.65-4.76 (m, 4 H), 4.96-5.06 (m, kis(dibenzy1phosphate) (31). Phosphorylation of (-)-30 (0.3 3H),7.25-7.35 (m, 15H);I3CNMR(CDCl3)672.38,73.12, g, 1 mmol), as described for 5, gave (+)-31(1.4 g, 98%): [a]D 73.86, 74.7, 75.06, 75.48, 77.24, 77.61, 81.01, 81.33 (2 C), + 8 S 0 (c 1.2, CHC13); 'H N M R (CDC13) 6 1.25-1.71 (m, 10 127,75(2C), 127.81 (2C), 127.85 (2C), 128.28 (2C), 128.55 H), 4.26 (t, 1 H, J = 5.88 Hz), 4.68 (dd, 1 H, J = 3.62,6.67 (2 C), 138.17 (2 C), 138.71 (2 C), 138.81. Hz), 4.70-4.80 (m, 2 H), 4.9 1-5.12 (m, 16 H), 7.18-7.29 (m, (+)-2,3,4-Tri- O-benzyl-myo-inositoIl,5,6Tris(dibenzy1 40 H). phosphate) (26). The triol (+)-25 (380 mg, 0.84 mmol) was 1 - D - myo- Inositol 3,4,5,6- Tetrakisphosphate [Ins(3,4,5,6)phosphorylated, as described for 5, to afford 26 (syrup, 0.9 &I. Hydrogenolysis of (+)-31(1.2 g, 0.9 mmol) followed by g, 92%): [ a ] +8.5' ~ (C 1.55, CHC13); 'H NMR (CDCl3) 6 KOH titration and lyophilization, as described for Ins(4,5)3.42 (dd, 1 H, J = 2.21, 5.96 Hz), 4.06 (t, 1 H, J = 9.6 Hz), P2, furnished Ins(3,4,5,6)P4 as the octapotassium salt (0.72 4.16-4.22 (m, 1 H), 4.40-4.49 (m, 4 H), 4.63-5.10 (m, 17 g, 99%): [ a ] +7.2' ~ (C 2.3, H2O); 'H N M R (D2O) 6 3.95 H), 6.97-7.35 (m, 45 H). (t, 1 H, J = 3 Hz), 4.04-4.10 (m, 2 H), 4.36 (br d, 2 H, J 1 -D-myo-Inositol1,5,6- Trisphosphate [Ins(1,5,6)Pj]. Hy= 10.2 Hz), 4.50 (br d, 1 H, J = 9 Hz); 31PN M R (D20, drogenolysis of (+)-26(0.9 g, 0.7 mmol) followed by KOH external H3P04) 6 3.32, 3.59, 3.78, 4.38. titration and lyophilization, as described for Ins(4,5)P2, gave (&)- 3,6-Di- 0-allyl- I ,2:4,5-di- 0-cyclohexylidene-myoIns(1,5,6)P3 as the hexapotassium salt (0.47 g, 99%): [a]D inositol (33). Allylation of racemic 1,2:4,5-di-O-cyclohex-3' (c 1.2, H2O); 'H N M R (D2O) 6 3.47-3.52 (m, 1 H), ylidene-myo-inositol(32) (1.5 g, 4.4 mmol), as described for 3.73-3.80 (m, 3 H), 3.83-3.89 (m, 1 H), 4.15-4.25 (m, 1 H), 21,provided (&)-33 (syrup, 1.75 g, 94%): 'H NMR (CDC13) 4.29 (t, 1 H, J = 2.83 Hz); 31PN M R (D20, external H3P04) 6 1.25-1.76 (m, 20 H), 3.31 (dd, 1 H , J = 9.2, 10.5 Hz), 3.63 6 3.56, 3.57, 4.45. (dd, 1 H , J = 6.5, 10.5Hz),3.76(dd, 1 H,J=4.1,25.2Hz), 3.97 (t, 1 H, J = 6.45 Hz), 4.06 (dd, 1 H, J = 4.95,6.60 Hz), (+)-3,4-Di-0-benzyl-myo-inositol (27). Hydrolysis of the cyclohexylidene rings of (+)-19 (0.7 g, 1.3 mmol), as 4.21-4.37 (m, 4 H), 4.45 (t, 1 H, J = 3.9 Hz), 5.15-5.36 (m, described for 14,generated (+)-27(0.46 g, 95%): [a]D +6' 1 H), 5.86-6.05 (m, 2 H). (C 0.5, CH30H); 'H N M R (CD30D) 6 3.41 (dd, 1 H, J = (&)- 3,6- Di- 0 - a l l y l - I ,2- 0-cyclohexylidene-myo-inosi2.71, 6.16 Hz), 3.66 (t, 1 H, J = 9.3 Hz), 3.74 (t, 1 H, J = tol (34). Selective hydrolysis of the trans-cyclohexylidene of 9 . 3 H z ) , 4 . 1 6 ( t , l H , J = 2 . 1 5 ) , 4 . 6 6 ( q , 2 H , J = 11.5,30.1 (&)-33(1 g, 2.3 mmol), as described for 4,gave (&)-34(solid, Hz), 4.95 (s, 4 H), 7.26-7.40 (m, 10 H); I3C NMR (CDCl3) 0.8 g, 98%): 'H NMR (CDC13) 6 1.26-1.78 (m, 10 H), 2.79 668.93,70.54,71.46,72.79,73.7,75.0,79.74,81.17,126.76, (br s, 2 H), 3.36 (t, 1 H, J = 9.36 Hz), 3.44-3.51 (m, 2 H), 126.97, 127.20 (2 C), 127.70 (2 C), 127.83 (2 C), 138.89, 3.90 (t, 1 H, J = 9.37 Hz), 4.06 (dd, 1 H, J = 5.09,5.95 Hz), 139.56. 4.16-4.30(m,3H),4.39-4.47(m,2H),5.17-5.25(m,2H), 5.26-5.36 (m, 2 H), 5.89-6.04 (m, 2 H); 13CN M R (CDC13) I -D-myo-Inositol 1,2,5,6-Tetrakisphosphate [Ins(l,2,5,6)623.61,23.92,25.01,35.24,37.71,71.29,71.52,72.34,73.21, PI]. Phosphorylation of (+)-27 (0.35 g, 1.4 mmol), as described for 5, furnished (+)-3,4-di-O-benzyl-myo-inositol 73.44, 77.26, 78.93, 82.13, 110.59, 117.08, 117.59, 134.85, tetrakis(dibenzy1phosphate)(+)-28(syrup, 1.1 g, 81%): [ a ] ~ 135.02. Anal. Calcd for C18H2sOa: C, 63.51; H, 8.29. Found: C, 63.6; H, 8.25. +5.8' (c 1.9, CHCl3). Hydrogenolysis of (+)-28followed by KOH titration and lyophilization, as described for Ins(4,5)(&)- 3,4,6- Tri- 0-allyl- I ,2- 0-cyclohexylidene- myo- inosiP2, afforded Ins(1,2,5,6)P4 (0.2 g, 99%): [a]D -4.9' ( c 1.3, tol ( 3 5 ) . Regioselective allylation of (f)-34(0.8 g, 2.3 mmol) H2O); 'H N M R (D20) 6 3.34 (dd, 1 H, J = 2.1, 5.2 Hz), with allyl bromide, BuzSnO, and CsF, as described for 10, 3.61-3.77 (m, 3 H), 4.21 (q, 1 H , J = 9.9, 18 Hz), 4.57-4.60 gave (&)-35 (syrup, 0.58 g, 65%): lH NMR (CDC13) 6 1.26-

Inositol Phosphate Recognition at IP3 Receptor

Biochemistry, Vol. 33, No. 38, 1994 11591

J = 9 Hz), 3.74-3.80 (m, 2 H), 4.08 (9,2 H, J = 9.3, 13.9 1.80 (m, 10 H), 2.70 (s, 1 H), 3.57 (dd, 1 H, J = 7.91, 9.72 Hz), 4.45 (t, 1 H, J = 1.25 Hz); 31PN M R (DzO, external Hz), 3.54-3.66 (m, 3 H), 4.06 (dd, 1 H, J = 5.69, 6.89 Hz), H3PO4) 6 3.23 (2 P), 4.05 (2 P). 4.18-4.28 (m, 3 H), 4.32-4.43 (m, 3 H), 5.12-5.21 (m,2 H), 5.26-5.27 (m, 2 H), 5.32-5.33 (m, 2 H), 5.87-6.03 (m, 3 H); (*)- 1,2- 0-Cyclohexylidene-myo-inositol(42).Selective 13CNMR (CDC13) 6 23.59,23.92,25.07, 35.1,37.28,72.08, hydrolysis of the trans-cyclohexylidene of (&)-32 (1.5 g, 4.4 72.29,73.23,73.38,74.17,77.20,78.54,80.13,81.45,110.51, mmol) with acetyl chloride in CH2C12-CH30H, as described 116.99, 117.22, 117.38, 134.93, 134.98 (2 C). for 4, gave (*)-42 (solid, 1 g, 82%): mp 173-174 OC; 'H (A)- 3,4,6- Tri- 0-allyl- 5- 0-benzyl- 1,2-0-cyclohexylideneNMR(CD3OD)61.47-1.71 (m,lOH),3.11 (t, 1 H , J = 9 . 3 myo-inositol(36). Benzylationof (f)-35(0.72g, 1.8 mmol), Hz), 3.47-3.68 (m, 3 H), 3.91 (dd, 1 H, J = 5.03, 7.46 Hz), as described for 3, afforded (&)-36 (syrup, 0.74 g, 81%): lH 4.35 (t, 1 H, J = 4.47 Hz); I3C N M R (CD3OD) 6 24.61, NMR (CDC13) 6 1.27-1.77 (m, 10 H), 3.30 (dd, 1 H, J = 24.94, 26.11, 36.21, 39.13, 71.8, 73.85, 75.14, 77.01, 77.35, 8.67-9.67 Hz), 3.56 (dd, 1 H, J = 6.18, 8.70 Hz), 3.62 (dd, 80.25,111.36. Anal. Calcdfor C12H2006: C, 55.37; H, 7.47. lH,J=7.20,8.65Hz),3.73(t,lH,J=8.40Hz),4.04(dd, Found: C, 55.17; H, 7.89. 1 H, J = 5.40, 5.47 Hz), 4.17-4.40 (m, 6 H), 4.77 (s, 2 H), (&)- 3,4,5,6- Tetra- 0-allyl- I ,2- 0-cyclohexylidene-myo5.13-5.21 (m, 3 H), 5.24-5.34 (m, 3 H), 5.88-6.04 (m, 3 H), inositol (43). Exhaustive allylation of (*)-42 (0.9 g, 3.2 7.27-7.40 (m, 5 H); 13CNMR(CDC13)6 23.58,23.91,25.07, mmol), as described for 21, provided (*)-43 (syrup, 1.3 g, 35.23, 37.37, 72.37, 72.98, 73.85, 74.20, 75.4, 77.22, 78.74, 92%); lH N M R (CDC13) 6 1.28-1.80 (m, 10 H), 3.16 (dd, 80.47, 82.21, 82.46, 110.47, 116.69, 116.71, 117.30, 127.61, 1 H, J = 8.75, 8.85 Hz), 3.49-3.59 (m, 2 H), 3.67 (t, 1 H, 128.14 (2 C), 128.31 (2 C), 135.11, 135.24, 135.37, 138.72. J = 8.75), 4.00 (dd, 1 H, J = 5.45, 6.25 Hz), 4.16-4.37 (m, (f)-3,4,6- Tri- 0-allyl-5- 0-benzyl-myo-inositol(37). Hy9 H), 5.12-5.20 (m, 4 H), 5.24-5.33 (m, 4 H), 5.87-6.03 (m, drolysis of the cis-ketal of (*)-36 (0.7 g, 1.48 mmol) with 4 H). acetyl chloride in CH2C12-CH30H, as described for 14, (&)- 3,4,5,6- Tetra- 0-allyl-myo-inositol(44).Hydrolysis furnished (f)-37 (syrup, 0.5 g, 86%): lH NMR (CDC13) 6 ofthecis-ketalof(f)-43(1.3 g, 3.2mmol) withacetylchloride 2.48-2.52 (br s, 2 H), 3.28-3.38 (m, 2 H), 3.43-3.49 (m, 1 in CH2C12-CH30H, as described for 14, furnished (&)-44 H), 3.62-3.76 (m, 2 H), 4.09-4.44 (m, 5 H), 4.81 (dd, 2 H, (syrup, 0.9 g, 85%): 'H NMR (CDC13) 6 2.65 (br s, 2 H), J = 10.7,15.4Hz),5.13-5.34(m,6H),5.86-6.02(m,3H),3.16-3.26 (m, 2 H), 3.40 (dd, 1 H, J = 2.7, 6.75 Hz), 3.567.26-7.37 (m, 5 H); 13CNMR (CDC13)6 69.49,71.75,71.84, 3.69(m,2H),4.16-4.44(m,9H),5.12-5.33 (m,8H),5.8574.35, 74.58, 75.71, 77.23, 79.7, 80.85, 81.27, 83.16, 93.03, 6.03 (m, 4 H). 116.61,117.19,117.67,128.02,128.39,134.67,135.08,135.33, I ,3,4,5,6-Penta- 0-allyl-myo-inositol(45). Regioselective 138.66. allylation of (&)-44 (0.85 g, 2.3 mmol) with allyl bromide, 1,3,4,6- Tetra- 0allyl- 5- 0-benzyl-myo-inositol(38). ReBuzSnO, and CsF, as described for 10, gave 45 (syrup, 0.9 g, gioselective allylation of (&)-37 (0.36 g, 0.92 mmol) with 95%): lH N M R (CDC13) 6 2.45 (br s, 1 H), 3.16-3.21 (m, allyl bromide, BuzSnO, and CsF, as described for 10, gave 3 H), 3.67 (t, 1 H , J = 9.3 Hz),4.15-4.19 (m, 6H),4.27-4.30 (*)-35 (syrup, 0.33 g, 84%): lH N M R (CDCl3) 6 2.65 (br (m, 6 H), 5.1 1-5.20 (m, 5 H), 5.23-5.33 (m, 5 H), 5.86-6.03 s, 1 H), 3.21-3.35 (m, 3 H), 3.73 (t, 1 H, J = 9.06 Hz), (m, 5 H); 13C N M R (CDCl3) 6 71.71 (2 C),74.0, 74.51 (2 4.16-4.19(m,3H),4.29-4.32(m,3H),4.81 (s,2H),5.12C), 74.61, 74.74, 80.45 (2 C), 81.33 (2 C), 83.32, 116.36 (4 5.34 (m, 10 H), 5.85-6.04 (m, 4 H), 7.27-7.38 (m, 5 H). C), 127.26, 127.92 (2 C), 128.07 (2 C), 135.12 (2 C), 135.61 1,3,4,6-Tetra- O-allyl-2,5-di- 0-benzyl-myo-inositol(39). (2 C), 135.67 (2 C), 139.13. Conventional benzylation of 38 (0.31 g, 0.72 mmol), as 1,3,4.5,6- Penta- 0-allyl-2- 0-benzyl-myo-inositol (46). described for 3, provided 39 (syrup, 0.36 g, 96%): 'H N M R Conventional benzylation of 45 (0.8 g, 2 mmol), as described (CDC13) 6 3.18 (dd, 2 H, J = 2.3, 9.8 Hz), 3.32 (t, 1 H, J = for 3, provided 46 (syrup, 0.8 g, 81%): lH N M R (CDCl3) 6 9.21 Hz), 3.82 (t, 2 H, J = 9.5 Hz), 3.96 (t, 1 H, J = 2.64 3.11-3.19(m,3H),3.75(t,2H,J=6,6Hz),3.94(t,lH, Hz), 4.06-4.10 (m, 4 H), 4.24-4.38 (m, 4 H), 4.81 (s, 2 H), J = 2.4 Hz), 4.01-4.13 (m, 4 H), 4.22-4.35 (m, 6 H), 4.83 4.82 (m, 1 H), 5.10-5.33 (m, 8 H), 5.83-6.03 (m, 4H), 7.24(s, 2 H), 5.10-5.17 (m, 5 H), 5.22-5.32 (m, 5 H), 5.82-6.03 7.41 (m, 10 H). (m, 5 H), 7.22-7.42 (m, 5 H); 13C NMR (CDCl3) 6 71.46 2,5-Di-O-benzyl-myo-inositol (40). Deallylation of 39 (2 C), 72.51 (2 C), 74.4, 75.37, 80.32, 127.92, 128.0 (2 C), (0.5 g, 0.95 mmol) with 10% Pd/C and p-toluenesulfonic 128.4 (2 C), 137.7. acid, as described for 17,furnished 40 (solid, 0.2 g, 58%): mp 2-O-Benzyl-myo-inositol(47). Deallylation of 46 (0.6 g, 270-272 "C (dec); 'H NMR (DMSO-d6) 6 3.02 (t, 1 H, J 1.2 mmol) with 10% Pd/C and p-toluenesulfonic acid, as = 9.0 Hz), 3.28-3.31 (m, 2 H), 3.54-3.62 (m, 2 H), 3.71 (t, described for 17, furnished 47 (solid, 0.28 g, 80%): mp 2481 H, J = 2.4 Hz), 4.73 (d, 2 H, J = 4.8 Hz), 4.76 (s, 4 H), 250 (dec); lH N M R (D20) 6 3.10 (t, 1 H, J = 8.5 Hz), 4.81 (d, 2 H, J = 4.8 Hz), 7.19-7.42 (m, 10 H); 13C NMR 3.49-3.61 (m, 4 H), 3.88 (t, 1 H, J = 2.6 Hz), 4.69 (s, 2 H), (CDC13) 6 72.07 (2 C), 72.87 (2 C), 73.4,73.96 (2 C), 81.44, 7.23-7.35 (m, 5 H). Anal. Calcd for C12H1606: C, 56.24; 83.98, 126.69 (2 C), 126.82 (2 C), 127.28 (2 C), 127.65 (2 H, 6.29. Found: C, 56.46; H, 6.47. C), 127.72 (2 C), 139.71. Anal. Calcd for C18H2006: C, 65.05; H, 6.06. Found: C, 65.48; H, 5.99. 2- 0-Benzyl- myo- inositol I ,3,4,5,6-Pentakis(dibenzy1phosphate) (48). The foregoing intermediate47 (0.2g, 0.69 mmol) 2,5-Di-0- benzyl-myo-inositol I ,3,4,6- Tetrakis(dibenzy1 was phosphorylated, as described for 5, to afford 48 (syrup, phosphate) (41). The tetrol 40 (0.2 g, 0.55 mmol) was 1 g, 91%): lH NMR (CDC13) 6 4.25-4.42 (m, 3 H), 4.73phosphorylated, as described for 5, to afford 41 (syrup, 0.7 4.75 (m, 3 H), 4.89-5.07 (m, 22 H), 7.13-7.25 (m, 5 5 H). g, 91%): 'H NMR (CDC13) 6 3.45-3.51 (m, 2 H), 4.21-4.28 (m, 2 H), 4.64-5.09 (m, 22 H), 7.08-7.28 (m, 50 H). myo- Inositol I ,3,4,5,6-Pentakisphosphate [Zns(1,3,4,5,6)myo-Inositol1,3,4,6- Tetrakisphosphate [Ins(l,3,4,6)Pd]. PSI. Hydrogenolysis of 48 (0.8 g, 0.5 mmol) followed by Hydrogenolysis of the foregoing intermediate 40 (0.6 g, 0.42 KOH titration and lyophilization, as described for Ins(4,5)mmol) followed by KOH titration and lyophilization, as P2, gave Ins( 1,3,4,5,6)P5 as the decapotassium salt (0.46 g, described for Ins(4,5)Pz, gave Ins( 1,3,4,6)P4 as the octapo99%): ' H NMR (D20) 6 3.96 (t, 1 H, J = 3.3 Hz), 4.05 (br tassium salt (0.34 g, 99%): lH N M R (D20) 6 3.39 (t, 1 H, d, 1 H, J = 9.6 Hz), 4.37 (br s, 1 H), 4.41 (br s, 2 H), 4.45

11592 Biochemistry, Vol. 33, No. 38, 1994 (t, 1 H, J = 2.7 Hz); 31PN M R (D20, external H3P04) 6 4.49 (2 P), 4.89 (1 P), 5.42 (2 P). Preparation of Crude Rat Brain Microsomes. Adult rat brain (250 g) was homogenized (Polytron, setting 7 and 5 strokes) in ice-cold buffer A, consisting of 0.32 M sucrose, 10 mM Tris-HC1 (pH 7.3), 1 mM dithiothreitol (DTT), and 0.5 mM p-toluenesulfonyl fluoride (PMSF). The homogenate was centrifuged at 500gat 4 OC for 20 min, and the supernatant was first centrifuged at 12OOOg at 4 "C for 20 min to remove the mitochondria and then pelleted by centrifugation at 35000g at 4 OC for 40 min. The pellet was resuspended in buffer B, consisting of 20 mM Hepes-KOH (pH 7.3), 150 mM KCl, 3 mM MgCl2, and 1 mM DTT. The washing procedure was repeated two times. The resulting final pellet was suspended in buffer B to a protein concentration of 2-3 mg/mL. Ca2+Release Assay. Measurements of free Ca2+concentrations in incubation media were performed by using a Ca2+sensitive fluorescent dye, Fura-2, in a Hitachi F-2000 spectrofluorimeter. The assay medium consisted of 40 units of creatine kinase, 20 mM creatine phosphate, 5 pg of oligomycin, and 0.5 pM-Fura 2 free acid in 2 mL of buffer B containing 0.2-0.25 mg of microsomal proteins. Themixture was incubated for 5 min and treated with 1 mM ATP to allow the loading of Ca2+stores. Until the external Ca2+concentration returned to a near-base level, the microsomes were stimulated with the Ca2+-mobilizing agents. Experiments were carried out at 37 OC. Excitation and emission wavelengths were 340 and 510 nm, respectively. [Ca2+]i was calculated according to the equation [Ca2+]i= &(F- Fmin)/ (Fmax - F'),where F,,, and Fmin are readings in the presence of 2.5 mM CaC12 and 1 mM EGTA, respectively; Kd denotes the apparent dissociation constant of the Ca2+-dyecomplex. The Fura-2 fluorescence ratio signal was calibrated as described by Grynkiewicz et al. (1985). Ins(1,4,5)Pj Receptor-Binding Assay. Rat cerebellar microsomes were prepared according to the preceding procedure and suspended at 2 mg/mL in 50 mM Tris-HC1 (pH 8.3), 100 mM KCl, 1 mM EDTA, and 0.25 mM DTT. The [3H]Ins(1,4,5)P3-binding assay was performed by the membrane filtration method (Spat et al., 1986; Worley et al., 1987), in which 50 pg of microsomes in 250 pL of the same buffer was incubated with 2 nM [3H]Ins( 1,4,5)P3and individual inositol phosphates at the indicated concentrations for 30 min at 4 "C.Nonspecific binding was measured in the presence of 10 pM cold Ins( 1,4,5)P3. Incubations were terminated by filtration through phosphate-presoaked GF/C filters (Whatman), followed by 3 X 3 mL washes with the ice-cold incubation buffer. The receptor-bound radioactivity was analyzed by liquid scintillation spectrometry.

RESULTS Systematic Synthesis of D-myo- Inositol Polyphosphates. Our key synthetic strategy focused on the preparation of a pair of enantiomerically active 1,2:5,6-dicyclohexylidene-myoinositols, (+)- and (-)-2, as common precursors. Both enantiomers have a wide scope of application and allow the synthesis of virtually all of the inositol phosphates and inositol phospholipids. Preparation of the pure enantiomers of 2 was achieved by a facile enzymatic resolution, shown in Scheme 2. The 4-butyryl monoester of 2, (*)-l, was subjected to enantiospecific hydrolysis by porcine pancreatic lipase in a biphasic mixture consisting of h e x a n e 4 1 M potassium phosphate buffer (pH 7.0). Enantiospecificity of the enzymatic hydrolysis was greater in the biphasic system than in aqueous solutions in part due to the inhibition of contaminating proteases and esterases by the organic phase. Both product

Lu et al. Scheme 2: Enzymatic Preparation of (+)- and (-)-2

t

"'OR

c = 0.5 0-1

ee(producl) = O.%*

ee(,,bs,rate)= 0.98* R = COC,H,, (-) 1 = H.(+)-2 (*after recrystallization)

(-1.2

2

(-)-2 and substrate (-)-1 fractions were obtained with satisfactory optical purity (ee = 0.98) after recrystallization. This expedient and inexpensive method permitted the preparation of both antipodes of 2 in good quantities. The synthetic versatility of 2 is underscored by the preferential removal of the trans-5,6-cyclohexylidenering vs the cis-ketal and the regiospecific distinction of the vicinal diols (1- vs 6-OH) via the corresponding stannylene acetal. Accordingly, the combination of selective protections and deprotections led to a variety of useful intermediates for the synthesis of target inositol phosphates, the details of which are described in the following sections. The syntheses of Ins(1,4,5)P3 from (+)-2 and of Ins(1,3,4)P3 and Ins(1,3,4,5)P4 from (-)-2 have been described elsewhere (Gou et al., 1992) and will not be elaborated here. Synthetic Utility of (+)-2. As shown in Scheme 3, (+)-2 provided a convenient route to both Ins(4,5)P2 and Ins( 1,4,5,6)P4. Exhaustive benzylation of (+)-2 in tandem with the selective hydrolysis of the trans-cyclohexylidene group afforded a key intermediate, (+)-4.Phosphorylation of the diol 4 by the phosphoramidite method, followed by debenzylation and removal of the cis-cyclohexylidene group, gave Ins(4,5)P2 with a 65% overall yield from (+)-2. Efficient use of (+)-2 to prepare Ins( 1,4,5,6)P4 was made by selectively removing the trans-cyclohexylidene via the 1,6-diacetyl ester (-)-6, followed by alkaline hydrolysis to yield (+)-8. Accordingly, the tetrol 8 was transformed into Ins(1,4,5,6)P4 [77% from (+)-21. Synthetic Utility of (-)-2. Retrosynthetically, Ins( 1,4)P2, Ins( 1,5,6)P3, Ins( 1,2,5,6)P4, and Ins(3,4,5,6)P4 could be obtained from (-)-2,as illustrated in Scheme 4. Regiospecific introduction of allyl and benzyl groups to the C-4 and C-3 hydroxyl functions, respectively, of (-)-2 gave the fully protected derivative (+)-11. Selective removal of the transketal, followed by exhaustive benzylation, afforded (-)-13, the methanolysis of which provided the diol (-)-14. 1-0Allylation of 14 yielded (+)-15,and subsequent benzylation at the 2-OH followed by deallylation with Pd/C and p-toluenesulfonic acid furnished a crucial diol, (+)- 17. Phosphorylation and debenzylation of 17 gave Ins( 1,4)P2 [21% from (-)-21. With regard to the synthesis of Ins(1,5,6)P3, (-)-2 was subjected to exhaustive benzylation to give (+)-19, and the selective hydrolysis of the trans-cyclohexylidene group yielded (-)-20. Di-0-allylation of 20 provided (-)-2l, and removal of the cis-ketal by methanolysis gave the diol (-)-22. 1-0-Allylation of 22 provided (-)-23, and benzylation at the 2-OH followed by deallylation furnished a pertinently protected triol (+)-25, which afforded Ins( 1,5,6)P3accordingly, with an overall yield of 39% from (-)-2. Alternatively, (+)19 could undergo exhaustive methanolysis to afford 3,4-di0-benzyl-D-myo-inositol (+)-27,which was then converted toIns(1,2,5,6)P4 [77%from(-)-2]. Theuseof (-)-2 toprepare Ins(3,4,5,6)P4 was made in a straightforward manner. To

Biochemistry, Vol. 33, No. 38, 1994 11593

Inositol Phosphate Recognition at IP3 Receptor Scheme 3: Synthetic Utility of (+)-2"

'//OB n OR R = -P(0)(OBn)2

'"'OBn

OH (-1-3

(-1.6

(+I4

(+)-4

OH

OH

(-)-7

(+)-8

OR R = -P(0)(OBn)2 (-1-9

Route A, Ins(4,5)Pz synthesis; route B, Ins( 1,4,5,6)P4synthesis.

Scheme 4: Synthetic Utility of

(-)-2O

OH

0

R = H, ( - ) - l o R = Bn, (+)-11

7

R i = H, (-)-12 R i = Bn, (-)-13

7

(-)-14

OR2

R2 = H, (+)-15 R2 = Bn, (+)-I6

7

R3 = H. (+)-17 R3 = P(0)(OBn)2, (+)-I8

J

OH

B OR

0-9 B n O A o

(+)-29 a

R = H,(-)-20

7

(-)-22

R = All, (-)-21

(-)-30

R'= H.(-)-23 R' = Bn, (+)-24

R2 = H. (+)-25 R2 =

R = P(0)(OBn)2 (+)-31

Route A, Ins(l,4)Pz synthesis; route B, Ins(1,5,6)P3synthesis; route C,Ins(1,2,5,6)P4synthesis; route D,Ins(3,4,5,6)P4synthesis.

prevent ketal migration, (-)-2 was subjected to acetylation to afford (+)-29. Selective removal of the trans-cyclohexylidene group furnished the tetrol (-)-30, the phosphorylation and then debenzylation of which provided the target molecule with an overall yield of 79% from (-)-2. Syntheses of Zns(1,3,4,6)Pd and Zns(1,3,4,5,6)P~. In view of the symmetric nature of Ins( 1,3,4,6)P4 and Ins( 1,3,4,5,6)P5, the syntheses did not call for the use of (+)- or (-)-2 as a chiral starting material. Thus, racemic 1,2:4,5-di-Ocyclohexylidene-myo-inositol(32), a major byproduct from the preparation of 2, was employed as a precursor to Ins(1,3,4,6)Pdand Ins(1,3,4,5,6)Ps (Scheme 5). Di-0-allylation of 32 in conjunction with the selective removal of the trans-

cyclohexylidene group gave the diol 34. Regiospecific allylation of 34 at the C-6 hydroxyl and subsequent benzylation led to the fully protected derivative 36, the methanolysis of which afforded 37. The diol 37 underwent regioselective allylation at the 1-OH position. This was followed by 2-0benzylation and subsequent deallylation to furnish the tetrol 40. Phosphorylation and then debenzylation of 40 provided Ins( 1,3,4,6)P4 (17% from 32). The pentakisphosphate Ins(1,3,4,5,6)P5 was synthesized in a similar manner. Selective removal of the trans-ketal of 32 followed by exhaustive allylation and subsequent methanolysis afforded the diol 44. Again, regiospecific introduction of allyl and benzyl groups to the hydroxyl functions at C-1 and (2-2, respectively, provided

11594 Biochemistry, Vol. 33, No. 38, 1994

04 .01

'

'

'

"""/

1

'

"

"

"

~

1

'

'

"""'

'

'

"""'

10

Lu et al.

'

100

'

'

"...I loo0

[Inositol polyphosphate] (pM)

FIGURE1: Inositol phosphate-induced Ca2+release from rat brain microsomes. CaZ+-loaded rat brain microsomes were treated with Ca2+-mobilizingagents, and the released Ca2+was monitored by a Caz+-sensitive fluorescent dye, Fura-2, according to the method described in the Experimental Procedures [loo% = Ca2+release at saturated concentrations of Ins( 1,4,5)P3].Each data point represents the mean of three determinations.

a fully protected derivative 46 which, after deallylation, led to 2-0-benzyl-myo-inositol (47). Phosphorylation and then debenzylation of 47 gave Ins( 1,3,4,5,6)P5 with an overall yield of 35% from 32. The chemical purity of these synthetic inositol phosphates was greater than 97% according to IH and 31P N M R spectroscopy. The amounts of isomeric impurities were negligible, as indicated by these N M R spectra. Inositol Phosphate- Induced ea2+Release from Rat Brain Microsomes. Ca2+-loadedrat brain microsomes were treated with individual synthetic D-myo-inositol phosphates at 37 "C, and the released Ca2+ was monitored by bulk fluorimetry using Fura-2 as an indicator. Of the 11 phosphoinositols examined, Ins( 1,4,5)P3, Ins( 1,3,4,6)P4, Ins( 1,3,4,5)P4, Ins(1,4,5,6)P4, and Ins(4,5)Pz exhibited Ca2+-mobilizingactivity in a dose-dependent manner (Figure l ) , with apparent EC50 values of 0.1 3,5, 13, 15, and 60 pM, respectively. It is worth noting that Ins( 1,3,4,6)P4, Ins( 1,3,4,5)P4, Ins( 1,4,5,6)P4,and Ins(4,5)P2 were capable of mobilizing the entire Ins( 1,4,5)P3-sensitive Ca2+store of rat brain microsomes, albeit at much lower potency. The Ca2+-mobilizingactivities of Ins( 1,3,4,6)P4, Ins( 1,3,4,5)P4, Ins( 1,4,5,6)P4, and Ins(4,5)P2 were 38-

fold, 100-fold, 1 15-fold, and 460-fold, respectively, lower than thatofIns(l,4,5)P3. Ins(1,3,4,6)P4(Ivorraetal., 1991;Gawler et al., 1991), Ins(1,3,4,5)P4 (Wilcox et al., 1993), and Ins(4,5)P2 (Burgess et al., 1984; Irvine et al., 1984) have been reported to be active in Ca2+ mobilization. Their reported potencies were in line with those of this study. Other inositol phosphates including Ins( 1,4)P2, Ins( 1,5,6)P3, Ins(1,3,4)P3, Ins(3,4,5,6)P4, Ins(1,2,5,6)P4, and Ins(1 ,3,4,5,6)P5 failed to exert appreciable Ca2+ release from the microsomal preparation, even at concentrations up to 100 PM. Inositol Phosphate Displacement of Specific [3H]Ins(1,4,5)Pj Binding in Rat Cerebellar Membranes. To assess the binding of inositol phosphates to the Ins( 1,4,5)P3 receptor, displacement of specific [3H]Ins(1,4,5)P3binding was carried out using rat cerebellar membrane preparations. Figure 2A,B illustrates the displacement curves for the inositol phosphates examined. Accordingly, the mean IC50 values for individual compounds were determined as follows: Ins( 1,4,5)P3, 0.03 1 pM; Ins( 1,4,5,6)P4, 1.7 pM; Ins( 1,3,4,5)P4, 1.8 pM; Ins(1,3,4,6)P4, 2.3 pM; Ins(4,5)Pz, 24.3 pM; Ins(1,3,4,5,6)P5, 41 pM; Ins(3,4,5,6)P4, 58 pM; Ins(1,2,5,6)P4, 63 pM; Ins(1,3,4)P3, 125 pM; Ins(l,4)P2, 315 pM; Ins(1,5,6)P3, 375 pM. For the inositol phosphates capable of effecting Ca2+ mobilization, the relative potency of inhibiting [3H]Ins(1,4,5)P3 binding to cerebellar membranes was in line with the order of the EC5o values. Those molecules that failed to elicit Ca2+ mobilization were weak ligands for the receptor, presumably through nonspecific electrostatic interactions. The IC50 values were 1500-12000-fold higher than that of Ins( 1,4,5)P3.

DISCUSSION The metabolic turnover of inositol phosphates in various cell types generates more than 20 different metabolites (Majerus et al., 1988; Fisher et al., 1992; Shears, 1992; Menniti et al., 1993). To date, studies on the biochemical relevance of these inositol phosphates have primarily focused on the roles of Ins( 1,4,5)P3and Ins( 1,3,4,5)P4inmodulating cytosolic [Ca2+](Berridge, 1993). However, evidenceis mounting that other inositol phosphates may also participate in important cellular functions. As part of our systematic effort to unveil the biological utility of these phosphoinositols, in this paper we examined the molecular recognition of endogenous inositol

Inositol Phosphate Recognition at IP3 Receptor 1

Ins( 1,4,5)P3 Ins( I ,3,4,5)P4

Ins( 1.4.5,6)P4 Ins( 1,3,4,6)P4 Ins(4.5)PL Ins( 1,3,4.5,6)P5

Ins( 1,2,5,6)P4 Ins(3,4,5,6)P4 lns(1,3,4)P3 Ins( 1,5,6)P3 Ins( 1,4)PL

2o

1

04

1

v '

'

'

"""I

IO

'

'

"""I

IO0

'

'"""I

loo0

'

"..-I loo00

[Inositol polyphosphate] (pM) FIGURE2: Inhibition of specific [3H]Ins(1,4,5)P3binding to rat cerebellar microsomes by increasing concentrations of inositol phosphates: (A) Ins(1,4,5)P,,Ins(1,3,4,5)P4,Ins( 1,4,5,6)P4,Ins( 1,3,4,6)P4, Ins(4,5)P2,and Ins( 1,3,4,5,6)P5;(B) Ins(1,2,5,6)P4,Ins(3,4,5,6)P4, Ins( 1,3,4)P3,Ins( 1,5,6)P3,and Ins( 1,4)P2.Nonspecific binding was measured in the presence of 10 WMcold Ins(1,4,5)P3.Each data point represents the mean of at least three determinations. phosphates at the Ins( 1,4,5)P3-specific receptor. Eleven inositol phosphate congeners, including Ins( 1,3,4,5,6)P5, Ins(1,2,5,6)P4,Ins( 1,3,4,5)P4, Ins( 1,3,4,6)P4,Ins( 1,4,5,6)P4, Ins(3,4,5,6)P4, Ins(1,3,4)P3, Ins(1,4,5)P3, Ins(1,5,6)P3, Ins(l,4)P2, and Ins(4,5)P2, were synthesized, many of which are unavailable from commercial sources. All but Ins( 1,5,6)P3 and Ins( 1,2,5,6)P4are naturally occurring compounds. These optically active inositol phosphates were independently synthesized from (+)- or (-)-2 in good yields and were fully characterized by IH and 31P NMR, with both chemical and optical purities greater than 97%. The present data clearly show that many of the inositol phosphates derived from Ins( 1,4,5)P3 metabolism are capable of eliciting intracellular Ca2+release, however, at lower potency than the parent compound. Of the 11 compounds examined, the Ca2+-releasing inositol phosphates include, in the order of ECso values, Ins( 1,4,5)P3, Ins( 1,3,4,6)P4, Ins( 1,3,4,5)P4, Ins( 1,4,5,6)P4, and Ins(4,5)P2. Binding experiments using rat cerebella membrane as a source of Ins( 1,4,5)P3-specific receptor further suggest that the ability of these molecules to elicit Ca2+mobilization arises from interactions with the Ins(1,4,5)P3-~pecificreceptor. As noted in Figure 2A, for these inositol phosphates, the range of concentrations for complete displacement of [3H]Ins( 1,4,5)P3 binding was approximately 1 log unit, which is a good indication of specific binding. Moreover, it is worth mentioning that Ins( 1,3,4,6)P4is a potent agonist at the Ins(1,4,5)P3 receptors of rat cerebella microsomes. The Ca2+-releasing activity of Ins( l ,3,4,6)P4 has also been observed in microinjected Xenopus oocytes (Ivorra et al., 1991) and in permeabilized human neuroblastoma cells (Gawler et al., 199 l), with potencies comparable to that found in this study (ECso = 5 pM). With the unique symmetric nature of the molecule, Ins( 1,3,4,6)P4may assume alternative binding conformations to the receptor, in which the important

Biochemistry, Vol. 33, No. 38, I994 11595 recognition features of Ins( 1,4,5)P3 can be mimicked (Potter & Nahorski, 1992; Mills et al., 1993). Previously, Gawler et al. (1991) indicated that DL-I~s( 1,4,5,6)P4 did not exhibit Ca2+-releasing activity at concentrations up to 10 pM. As mentioned earlier, racemic Ins( 1,4,5,6)P4 is a mixture of Ins(1,4,5,6)P4 and Ins(3,4,5,6)P4. In view of the facts that the latter compound is inactive in Ca2+ mobilization and that Ins( 1,4,5,6)P4has an ECsovalueof 15fiM, the Ca2+-mobilizing activity of the racemic mixture at 10 pM might not be appreciably noticed. The structureactivity relationships between inositol phosphates and the Ins( 1,4,5)P3 recognition sites have been the focus of discussion in many papers (Kozikowski et al., 1990, 1993a,b; Safrany et al., 1992a,b; Potter & Nahorski, 1992; Mills et al., 1993). It has generally been perceived that the vicinal 4,5-bisphosphate motif assumes an obligatory role in effecting productive binding with the Ins( 1,4,5)P3 receptor. The current study indicates that the absence of either the 4or 5-phosphate results in the complete loss of agonist property, as evidenced by the lack of activity in Ins( 1,4)P2, Ins( 1,3,4)P3, Ins( 1,5,6)P3,and Ins( 1,2,5,6)P4even at concentrations up to 100 pM. The importance of the 1-phosphate is implied by the potency of Ins(4,5)P2 relative to that of Ins( 1,4,5)P3. Comparison of the Ca2+ release and receptor-binding data indicates that Ins(4,5)P2 is a 460-fold weaker agonist and a 800-fold weaker ligand than Ins( 1,4,5)P3. Moreover, it is noteworthy that structural variants of Ins( 1,4,5)P3 with modification of the phosphate position, such as Ins(2,4,5)P3 (Burgess, 1984; Loomis-Husselbee & Dawson, 1993), Ins(1:2cyc,4,5)P3 (Wilson et al., 1985; Irvine et al., 1986), and Ins(1,2,4,5)P4 (Mills et al., 1993b; Hirata et al., 1994), were reported to retain full agonist and ligand properties. These results suggest that the 1-phosphate group contributes substantially to the binding through charge-charge interactions and/or hydrogen bonding with the binding domain. The ionic interaction is thought to be of a long-range nature to account for the relaxed stereochemical and positional specificities in recognizing this phosphate moiety. Previously, Kozikowski et al. (1993a) proposed that the vicinal 2,3-dihydroxyl functions played a nonessential role in receptor recognition on the basis of the finding that ~-3-deoxy-, ~-2,3-dideoxy-,and ~-3-deoxy-3-fluoro-Ins( 1,4,5)P3exhibited very potent ligand and agonist properties. Nevertheless, the present evidence indicates that Ins( 1,3,4,5)P4 is a much weaker agonist than Ins( 1,4,5)P3. Presumably, electrostatic repulsion of the three adjacent phosphate functions alters the conformation of the 4,5-bisphosphate motif, which weakens the binding affinity. In contrast, the contribution of the 6-hydroxyl function appears to be prominent. One result from the work of Kozikowski et al. (1993a,b) on ~-2,3,6-trideoxy-Ins(1,4,5)P3 supports this premise, as it was shown to be a much weaker agonist than 2,3-dideoxy-Ins( 1,4,5)P3. Also, we previously prepared 6-0-(w-aminohexyl)-Ins( 1,4,5)P3for the preparation of anti-Ins(l,4,5)P3 antibodies (Gou et al., 1994). The Ca2+mobilizing activity of this 6-0-substituted molecule was about 200-fold less than that of the parent molecule. Structural analysis indicates that all Ca2+-mobilizinginositol phosphates, including Ins( 1,3,4,6)P4and Ins( 1,4,5,6)P4, mimic or display a structure similar to that of the 4,5-bisphosphate 6-hydroxy motif of Ins(1,4,5)P3 (Figure 3). On the basis of these findings, we propose a binding model to account for ligand recognition at the Ins( 1,4,5)P3 receptor (Figure 4). The binding site of the Ins(1,4,5)P3 receptor is assumed to be composed of two domains. The anchoring domain interacts with the 4,5-bisphosphate 6-hydroxy motif, attributing to the Ca2+-mobilizingactivity. Disruption of this

11596 Biochemistry, Vol. 33, No. 38, 1994

Lu et al.

OH Ins( 1,4.5)P3 (0.13 pM; rat brain microsomes)

5

L-chiro-Ins(2$,5)P3 (1.4PM, neuroblastoma cells) (Safrany et al., 1992)

HO

Ins(4,5)P2 (60 WM,rat brain microsomes)

' - 0 - j m WOP032'-O~PO HO Ins( 1 ,2,4,5)P4 5

Adenine

om,2HO OH

Ins( 1.3,4,6)P4 (5 pM. rat brain microsomes)

(0.17 pM, neuroblastoma cells) (Mills et al., 1993)

0

2-0,po

opo,2-

HO Adenophostin A (1.5 nM, rat cerebellar microsomes) (Takahashi et al., 1994)

5

HO

5

HO

*-03mw Ins(2,4,5)P3 ( 1.3 pM,hepatocytes) (Burgess et al., 1984)

Ins( 1.3,4,5)P4 (13 pM,rat brain microsomes)

OH

2

-

~

3

~

OP032-

2

-

0 '3 -O,PW O

2-0,po

Ho

HO OH Ins( 1,4,6)P3 (2 pM, rabbit platelets) (Mills et al., 1993)

M

OH

Ins( 1 ,4,5,6)P4 (15 pM, rat brain microsomes)

FIGURE3: Structures of Caz+-mobilizinginositol phosphates and adenophostin A. All of these molecules assume conformations sharing or mimicking the structural feature (marked in bold) of the 4,s-bisphosphate 6-hydroxy and 1-phosphate motifs of Ins( 1,4,5)P3.Indicated in the parentheses are the EC50value and the source of Ins( 1,4,5)P3 receptor for the binding assay. nonessential for receptor binding

7

anchoring domain for the 4,5-bisphosphate6-hydroxy motif

auxiliary domain for long-range interaction with the 1-phosphate

FIGURE 4: Ligand recognition at the Ins( 1,4,5)P3-bindingsite. The 4,s-bisphosphate6-hydroxyand 1-phosphate motifs (marked in bold) of Ins( 1,4,5)P3 represent structural features essential to its Ca2+mobilizing and binding properties. Evidence suggests that the 2- and 3-OH functions do not have a crucial role in receptor binding. structural feature, especially the bisphosphate moiety, significantly diminishes or even abolishes the agonist activity. The auxiliary domain exerts long-range electrostatic interactions with the 1-phosphate group, which enhances the binding affinity. The stereochemical requirement for this phosphate recognition appears to be less stringent. As mentioned earlier, structural variants with modification of the position and stereochemistry of the 1-phosphate, such as Ins(2,4,5)P3, still retain full agonist activity. Moreover, due to the symmetric nature of the myo-inositol ring and the nonessential role of the 2- and 3-hydroxyl functions in receptor binding, inositol phosphates are able to assume alternative conformations, some of which mimic those structural motifs essential for productive binding. This supposition is supported by the Ca2+-releasing activityof Ins(l,3,4,6)PdandIns(1,4,6)P3 (Millset al., 1993a). Comparison of the structures of both molecules to that of Ins( 1,4,5)P3 (Figure 3) indicates that the 1,6-bisphosphate 5-hydroxy structure resembles the 4,s-bisphosphate 6-hydroxy

motif, while the 4-phosphate corresponds to the 1-phosphate. On the basis of a similar analogy, one would also expect Ins( 1,3,6)P3 to be a partial agonist. The proposed model accounts for the high potency of adenophostins (Figure 3) in Ca2+ mobilization, which was reported recently by Takahashi et al. (1994). Adenophostin A and B, metabolites isolated from Penicillium brevicompactum, showed Ca2+-releasingand receptor-binding activities 100-fold stronger than those of Ins( 1,4,5)P3. Although adenophostins and inositol phosphates are structurally unrelated molecules, one may envisage that these phosphoglucose derivatives display structural features for Ins( 1,4,5)P3 receptor recognition. Moreover, the fact that the auxiliary domain accommodates the bulky 2'-phosphoadenosine moiety suggests the size of the Ins( 1,4,5)P3-binding cavity. This information lends support to our supposition that the 1-phosphate of the bound Ins( 1,4,5)P3exerts a long-range ionic interaction with the receptor. For adenophostins, tight binding due to close proximity between the 2'-phosphate and the charged group on the receptor may account for their strong agonist activity. It is also worth noting that the removal of the 2'-phosphate function on the ribose ring of adenophostin A attenuated the activity by 1000-fold, which is in line with the relative potency of Ins(4,5)P2 to Ins(1,4,5)P3 (Takahsshi et al., 1994). Heparin (Ghosh et al., 1988) and related sulfated polysaccharides and decavanadate (Foehr et al., 1989) are known to competitively inhibit Ins( 1,4,5)P3 binding to its receptor. However, none of these antagonists is specific. It is likely that these negatively charged molecules gain access to the receptor through nonspecific electrostatic interactions with the binding domains, thus blocking the access of Ins( 1,4,5)P3 to the pocket.

Inositol Phosphate Recognition at IP3 Receptor In summary, we have demonstrated that many intracellular inositol phosphates are capable of effecting Ca2+mobilization by interacting with Ins( 1,4,5)P3-~pecificreceptors. At present, the biochemical implication of this observation remains to be explored since information on their intracellular distribution and concentrations as well as susceptibility to agonist stimulation is still lacking. One school of thought is that these inositol phosphates exert a concerted, although redundant, effort to maintain cytosolic [Ca2+]. However, one can argue that the Ca2+-mobilizing activity of these molecules is coincidental due to the sharing of common structural motifs. Several lines of evidence have indicated that some of these polyphosphates may be involved in cellular functions distinct from Ca2+ mobilization. Thus, further examinations of the biochemical functions of these intracellular inositol phosphates constitute the focus of this investigation.

REFERENCES Barker, C. J., Wong, N. S., Maccallum, S. H., Hunt, P. A., Michell, R. H., & Kirk, C. J. (1992) Biochem. J. 286, 469474. Berridge, M. J. (1993) Nature 361, 315-325. Bruzik, K. S., & Tsai, M.-D. (1992) J. Am. Chem. SOC.114, 6361-6374. Burgess, G. M., Irvine, R. F., Berridge, M. J., McKinney, J. S., & Putney, J. W., Jr. (1984) Biochem. J . 224, 741-746. Craxton, A., Erneux, C., & Shears, S. B. (1994) J . Biol. Chem. 269, 4337-4342. Davis, F. B., Davis, P. J., Lawrence, W. D., & Blas, S. D. (1991) FASEB J . 5, 2992-2995. Ely, J. A., Hunyard, L., Baukal, A. J., & Catt, K. J. (1990) Biochem. J. 268, 333-338. Ferris, C. D., & Snyder, S. H. (1992) in Advances in Second Messenger and Phosphoprotein Research (Vol. 26) (Putney, J. W., Jr., Ed.) pp 95-107, Raven Press, New York. Fisher, S. K., Heacock, A. M., & Agranoff, B. W. (1992) J . Neurochem. 58, 18-33. Foehr, K. J., Scott, J., Ahnert-Hilger, G., & Gratzl, M. (1989) Biochem. J. 262, 83-89. Garegg, P. J., Iversen, T., Johansson, R., & Lindberg, B. (1984) Carbohydr. Res. 130, 322-326. Gawler, D. J., Potter, B. V. L., Gigg, R., & Nahorski, S. R. (1991) Biochem. J . 276, 16-167. Ghosh, T. K., Eis, P. S., Mullaney, J. M., Ebert, C. L., & Gill, D. L. (1988) J . Biol. Chem. 263, 11075-11079. Gou, D.-M., & Chen, C.-S. (1992) Tetrahedron Lett. 33,721724. Gou,D.-M., Liu, Y.-C., & Chen, C.-S. (1992) Carbohydr. Res. 234, 51-64. Gou,D.-M.,Shieh, W.-R.,Lu,P.-J., & Chen,C.-S. (1994) Bioorg. Med. Chem. 2, 7-13. Grynkiewicz, G., Poenie, M., & Tsien, R. Y. (1985) J. Biol. Chem. 260, 3440-3450. Guse, A. H., Roth, E., & Emmrich, F. (1992) Biochem. J . 288, 489-495. Hirata, M., Narumoto,N., Watanabe, Y., Kanematsu, T., Koga, T., & Ozaki, S. (1994) Mol. Pharmacol. 45, 271-276. Hoer, A., & Oberdisse, E. (1991) Biochem. J. 278, 219-224. Irvine, R. F. (1992a) FASEB J. 6, 3085-3091. Irvine, R. F. (1992b) in Advances in Second Messenger and Phosphoprotein Research (Vol. 26) (Putney, J. W., Jr., Ed.) pp 161-185, Raven Press, New York. Irvine, R. F., Brown, K. D., & Berridge, M. J. (1984) Biochem. J. 222, 269-272. Irvine, R. F., Letcher, A. J., Lander, D. J., & Berridge, M. J. (1986) Biochem. J . 240, 301-304. Ivorra, I., Gigg, R., Irvine, R. F., & Parker, I. (1991) Biochem. J . 273, 317-321. Jenkinson, S., Challiss, R. A. J., & Nahorski, S. R. (1992) Biochem. J . 287, 437-442.

Biochemistry, Vol. 33, No. 38, 1994 11597

Kozikowski, A. P., Fauq, A. H., Aksoy, I. A., Seewald, M. J., & Powis, G. (1990) J. Am. Chem. SOC.112,7403-7404. Kozikowski, A. P., Ognyanov, V. I., Fauq, A. H., Nahorski, S. R., & Wilcox, R. A. (1993a) J . Am. Chem. SOC.115,44294434. Kozikowski, A. P., Ognyanov, V. I., Chen, C., Fauq, A. H., Safrany, S. T., Wilcox, R. A., & Nahorski, S. R. (1993b) J . Med. Chem. 36, 3035-3038. Kuo, T. H. (1988) Biochem. Biophys. Res. Commun. 152,11111116. Lin, G., Bennett, B., & Tsai, M.-D. (1990) Biochemistry 29, 2747-2757. Liu, Y.-C., & Chen, C . 3 . (1989) Tetrahedron Lett. 30, 16171620. Loomis-Husselbee, J. W., & Dawson, A. P. (1993) Biochem. J. 289, 861-966. Liickhoff, A,, & Clapham, D. E. (1992) Nature 355, 356-358. Majerus, P. W., Connolly, T. M., Bansal, V. S., Inhorn, R. C., Ross, T. S., & Lips, D. L. (1988) J . Biol. Chem. 263, 30513054. Marshall, I. B., & Taylor, C. W. (1993) J. Exp. Biol. 184, 161182. Mattingly, R. R., Stephens, L. R., Irvine, R. F., & Garrison, J. C. (1991) J. Biol. Chem. 266, 1514615153. Mayr, G. W. (1989) Biochem. J . 259, 463-470. Menniti, F. S., Oliver, K. G., Putney, J. W., Jr., & Shears, S. B. (1993) Trends Biochem. Sci. 18, 53-56. Mills,S. J., Al-Hafidh, J., Westwick, J., & Potter, B. V. L. (1993a) Bioorg. Med. Chem. Lett. 3, 2599-2604. Mills, S. J., Safrany, S. T., Wilcox, R. A., Nahorski, S. R., & Potter, B. V. L. (1993b) Bioorg. Med. Chem. Lett. 3, 15051510. Morris, A. P., Gallacher, D. V., Irvine, R. F., & Petersen, 0. H. (1987) Nature 330, 653-655. Parker, I., & Ivorra, I. (1991) J . Physiol. (London) 433, 207227. Potter, B. V. L. (1990) Nat. Prod. Rep., 1-24. Potter, B. V. L., & Nahorski, S. R. (1992) Biochem. SOC.Trans. 20, 434-442. Prestwich, G. D., Marecek, J. F., Mourey, R. J., Theibert, A. B., Ferris, C., Danoff, S. K., & Snyder, S. H. (199 1) J . Am. Chem. SOC.113, 1822-1825. Putney, J. W., Jr. (1992) in Advances in Second Messenger and Phosphoprotein Research (Vol. 26) (Putney, J. W., Jr., Ed.) pp 143-160, Raven Press, New York. Putney, J. W., Jr., Takemura, H., Hughes, A., Horstman, D. A., & Thastrup, 0. (1989) FASEB J. 3, 1899-1905. Reitz, A. B. (Ed.) (1991) Inositol Phosphates and Derivatives. Synthesis, Biochemistry and Therapeutic Potential, ACS Symposium Series 463, American Chemical Society, Washington, DC. Safrany, S. T., Wilcox, R. A., Liu, C.,Potter, B. V. L., & Nahorski, S.R. (1992a) Eur. J . Pharmacol. 226, 265-272. Safrany, S. T., Sawyer, D. A., Nahorski, S. R., & Potter (1992b) Chirality 4, 415-422. Shears, S. B. (1992) in Advances in Second Messenger and Phosphoprotein Research (Vol. 26) (Putney, J. W., Jr., Ed.) pp 63-92, Raven Press, New York. Smith, P. M. (1992) Biochem. J . 283, 27-30. Spat, A., Fabiato, A,, & Rubin, R. P. (1986) Biochem. J. 233, 929-9 32. Takahashi, M., Tanzawa, K., & Takahashi, S. (1994) J . Biol. Chem. 269, 369-372. Wilcox, R. A., Challiss, R. A. J., Baudin, G., Vasella, A., Potter, B. V. L., & Nahorski,S. R. (1993) Biochem. J. 294,191-194. Wilson, D. B., Connolly, T. M., Bross, T. E., Majerus, P. W., Sherman, W. R., Tyler, A. N., Rubin, L. J., & Brown, J. E. (1985) J. Biol. Chem. 260, 13496-13501. Worley, P. F., Baraban, J. M., Supattapone, S., Wilson, V. S., & Snyder, S. H. (1987) J . Biol. Chem. 262, 12132-12136.